LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Vietnamese Llama2 7B 40GB GPTQ by TheBloke

 »  All LLMs  »  TheBloke  »  Vietnamese Llama2 7B 40GB GPTQ   URL Share it on

  8   

Vietnamese Llama2 7B 40GB GPTQ is an open-source language model by TheBloke. Features: 7b LLM, VRAM: 4.1GB, Context: 4K, License: other, Quantized, LLM Explorer Score: 0.1.

  4-bit Base model:bkai-foundation-mod... Base model:quantized:bkai-foun...   Dataset:mc4   Dataset:pg19   Dataset:vietgpt/wikipedia vi   Dataset:wikipedia   En   Gptq   Llama   Quantized   Region:us   Safetensors   Vi
Model Card on HF 🤗: https://huggingface.co/TheBloke/vietnamese-llama2-7B-40GB-GPTQ 

Vietnamese Llama2 7B 40GB GPTQ Benchmarks

LLME Score: 0.10253
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Vietnamese Llama2 7B 40GB GPTQ (TheBloke/vietnamese-llama2-7B-40GB-GPTQ)
🌟 Advertise your project 🚀

Vietnamese Llama2 7B 40GB GPTQ Parameters and Internals

Model Type 
llama
Use Cases 
Areas:
General AI, Language Processing
Limitations:
Model requires further supervised fine-tuning (SFT) for practical applications.
Considerations:
Usage should comply with local laws and Meta's Llama-2 open-source license.
Additional Notes 
The new tokenizer reduces tokens by 50% compared to ChatGPT and approximately 70% compared to the original Llama2.
Supported Languages 
vi (native), en (supplementary)
Training Details 
Data Sources:
NewsCorpus, Vietnamese Wikipedia, Vietnamese books, Vietnamese legal documents, C4-vi, pg19, English Wikipedia, C4-en
Data Volume:
40.5 GB
Methodology:
single-epoch continual pretraining (incremental pretraining)
Training Time:
1000 GPU hours on DGX A100 with four GPU A100
Hardware Used:
DGX A100, four GPU A100
Model Architecture:
Vietnamese-specific vocabulary and token encoding enhancements, based on Llama2-chat 7B
Input Output 
Input Format:
{prompt}
LLM NameVietnamese Llama2 7B 40GB GPTQ
Repository 🤗https://huggingface.co/TheBloke/vietnamese-llama2-7B-40GB-GPTQ 
Model NameVietnamese Llama2 7B 40GB
Model CreatorBKAI-HUST Foundation Models Lab
Base Model(s)  bkai-foundation-models/vietnamese-llama2-7b-40GB   bkai-foundation-models/vietnamese-llama2-7b-40GB
Model Size7b
Required VRAM4.1 GB
Updated2026-03-29
MaintainerTheBloke
Model Typellama
Model Files  4.1 GB
Supported Languagesvi en
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length4096
Model Max Length4096
Transformers Version4.34.1
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size46303
Torch Data Typefloat16

Best Alternatives to Vietnamese Llama2 7B 40GB GPTQ

Best Alternatives
Context / RAM
Downloads
Likes
Yarn Llama 2 7B 128K GPTQ128K / 3.9 GB87
Yarn Llama 2 7B 64K GPTQ64K / 3.9 GB111
... 7B 32K Instructions V4 Marlin32K / 4.1 GB80
Aixcoder 7B GPTQ32K / 4.5 GB61
Calm2 7B Chat GPTQ32K / 4.4 GB25
...Calm2 7B Chat GPTQ Calib Ja 1K32K / 4.4 GB85
Llama 2 7B 32K Instruct GPTQ32K / 3.9 GB3127
Codebear 7B 4bit16K / 3.9 GB21
CodeLlama 7B Instruct GPTQ16K / 3.9 GB133346
...a 7B Instruct GPTQ Calib Ja 1K16K / 3.9 GB60
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/vietnamese-llama2-7B-40GB-GPTQ.

Rank the Vietnamese Llama2 7B 40GB GPTQ Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 52473 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer